Portability of semantic and spatial–temporal machine learning methods to analyse social media for near-real-time disaster monitoring

Natural Hazards - Up-to-date information about an emergency is crucial for effective disaster management. However, severe restrictions impede the creation of spatiotemporal information by current...     

Homographic motions of a Newtonian system of point masses

A complete classification is established of the possible types of homographic motions of a Newtonian system of point masses interacting through two-body forces which are arbitrary functions of their mutual separations. Several types of motion are included which had not been considered previously, but not all possible rigid body motions; thus it is not possible to build up rigid body dynamics from point dynamics by means of static two-body forces. All general results hold even if many-body forces are considered, which are functions of all separations. In an Appendix, the problem of homographic motions of a relativistic system of interacting point masses is outlined. For a particular form of relativistic interactions, the existence of several types of such motions is established.

---

Zusammenfassung Für ein Newtonsches System von Punktmassen mit Zweikörperkräften, die willkürliche Funktionen der Abstände sind, wird eine vollständige Klassifikation aller möglichen Ähnlichkeitsbewegungen gegeben. Sie schliesst mehrere Typen von Bewegungen ein, die bisher nicht in Betracht gezogen wurden, enthält aber nicht alle möglichen Starrkörperbewegungen; daher ist es nicht möglich, die Starrkörperdynamik von der Punktdynamik mit statischen Zweikörperkräften her aufzubauen. Alle allgemeinen Ergebnisse gelten selbst wenn Vielkörperkräfte in Betracht gezogen werden, die Funktionen aller Abstände sind. In einem Anhang wird das Problem der Ähnlichkeitsbewegungen eines relativistischen Systems von Punktmassen skizziert. Für eine besondere Form der relativistischen Wechselwirkung wird die Existenz mehrerer Typen solcher Bewegungen bewiesen.

---

Sommaire On établit une classification complète des types possibles des mouvements homographiques pour un système Newtonien de masses ponctuelles avec des forces entre paires de particules qui sont des fonctions arbitraires de leur distance mutuelle. Plusieurs types de mouvements sont inclus qui n'ont pas été considérés auparavant, sans toutefois inclure tous les mouvements possibles d'un corps solide; donc il n'est pas possible de construire la dynamique du corps solide en partant de la dynamique des points avec des forces statiques entre pairess de particles. Tous les résultats généraux restent valables même avec des forces généralisées dependant de toutes les distances. Dans un Appendice, on donne une prësentation générale du problème des mouvements homographiques pour un système relativiste de masses ponctuelles. L'existence de plusieurs types de tels mouvements est établie pour une forme particulère des interactions relativistes.

---

---

Research supported in part by the National Science Foundation.     

Combining machine-learning topic models and spatiotemporal analysis of social media data for disaster footprint and damage assessment

Current disaster management procedures to cope with human and economic losses and to manage a disaster’s aftermath suffer from a number of shortcomings like high temporal lags or limited temporal and spatial resolution. This paper presents an approach to analyze social media posts to assess the footprint of and the damage caused by natural disasters through combining machine-learning techniques (Latent Dirichlet Allocation) for semantic information extraction with spatial and temporal analysis (local spatial autocorrelation) for hot spot detection. Our results demonstrate that earthquake footprints can be reliably and accurately identified in our use case. More, a number of relevant semantic topics can be automatically identified without a priori knowledge, revealing clearly differing temporal and spatial signatures. Furthermore, we are able to generate a damage map that indicates where significant losses have occurred. The validation of our results using statistical measures, complemented by the official earthquake footprint by US Geological Survey and the results of the HAZUS loss model, shows that our approach produces valid and reliable outputs. Thus, our approach may improve current disaster management procedures through generating a new and unseen information layer in near real time.